Genotyping

DNA was prepared from tail biopsies or embryonic tissues using standard procedures. Genotyping of Tcfap2c alleles was done as described previously [12]. Primers used to detect the cre alleles were 5′-CCA CGA CCA AGT GAC AGC AAT G-3′ and 5′-CAG AGA CGG AAA TCC ATC GCT C-3′, both of which amplify a 373bp DNA fragment. Cycle conditions were 94°C for 45 sec, 56°C for 30 sec, and 72°C for 45 sec for 35 cycles.

Whole Mount Alkaline Phosphatase Staining of Embryos

Embryos were dissected in PBS, fixed in 4% (w/v) paraformaldehyde (PFA) in PBS for 10 min at room temperature, washed twice in PBS, and then incubated in freshly prepared alkaline phosphatase (AP)-staining solution, i.e., 250 mM Tris-maleic acid, pH 9, 0.4 mg/ml α-naphtylphosphate (Sigma-Aldrich, Taufkirchen, Germany), and 1 mg/ml Fast Red (Sigma-Aldrich), for 30 min in the dark. Embryos were cleared in 40% glycerol for 1 h and 80% glycerol overnight at 4°C. Embryos were photographed using a binocular (Leica, Bensheim, Germany) and a Zeiss Axiocam digital camera system (Carl Zeiss, Jena, Germany). Files were processed using Adobe Photoshop Elements and Illustrator software (Adobe, San Jose, CA).

Animal Care and Husbandry

All animals were housed at 25°C with a 12L:12D photoperiod and were given water and a standard rodent diet ad libitum. All animal experiments were approved by the Landesamt fuer Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (#8.87–50.10.31.08). The experiments were conducted in accordance with the International Guiding Principles for Biomedical Research Involving Animals as promulgated by the Society for the Study of Reproduction. To obtain Tcfap2c-deficient mice, we carried out a multistep mating scheme where mice heterozygous for the Tcfap2c allele (Tcfap2c+/−) were crossed with cre-expressing mice (cre+). The resulting Tcfap2c heterozygous cre double transgenic animals (Tcfap2c+/−; cre+) were crossed with Tcfap2c ^flox/flox mice to receive the null mutants. The day on which a copulation plug was found was defined as E 0.5.

Laser Microdissection

P.A.L.M. membrane slides (Carl-Zeiss MicroImaging, Göttingen, Germany) were irradiated with ultraviolet light for 30 min. Paraffin-embedded tissue was sectioned to 10 μm thickness and stained with haematoxylin and eosin using standard procedures. Seminiferous tubules were dissected using P.A.L.M. Microlaser Technology (Carl-Zeiss MicroImaging). Dissected tissue was collected in DNA lysis buffer (1 mM EDTA pH 8, 10 mM Tris-HCl pH 8, 0.5% Tween20, and 100 μg/ml proteinase K) and incubated at 37°C for 24 h, heated to 95°C for 12 min the next day, and used as a template for PCR.

RT-PCR

Total RNA was extracted using the Qiagen RNeasy Kit (Qiagen, Hilden, Germany); 100 ng was used for cDNA synthesis (SuperScriptIII; Invitrogen, Karlsruhe, Germany). PCR was performed with the primers listed in Supplemental Table S1 (all the Supplemental Data are available online at www.biolreprod.org). Thirty-five cycles (30 in the case of β-actin) were run using a standard protocol.

Quantitative RT-PCR

High-quality total RNA was extracted from the cell lines using an RNAqueous-4PCR kit (Ambion/Applied Biosystems, Nieuwerkerk a/d IJssel, the Netherlands) according to the manufacturer's instructions. Samples were treated with DNase and checked for residual DNA contamination by PCR using a primer set for SOX2, which is a single exon gene (see Supplemental Table S1). Complementary DNA was generated as described previously [17]. Quantitative PCR was performed using the 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA). For the detection of expression of a set of genes related to various differentiation lineages, a number of primer sets were used. Primer sequences are given in Supplemental Table S1. Target mRNA was quantified relative to HPRT transcripts as follows: target mRNA value = 2^{(mean Ct HPRT − Ct target)} [17].

Whole Mount In Situ Hybridization

Whole mount in situ hybridizations were performed as described previously [18]. Tcfap2c-full length cDNA subcloned in pBS vector was used for generation of the probe. Embryos were photographed using a binocular (Leica) and a Zeiss Axiocam digital camera system (Carl Zeiss). Files were processed using Adobe Photoshop Elements and Illustrator software (Adobe).

Histology and Imunohistochemistry

Embryos or organs were fixed in 4% neutral buffered formalin at 4°C overnight and embedded in paraffin. Sections of 1–3 μm thickness were stained with hematoxilin and eosin using standard procedures. For immunohistochemistry, sections were incubated using the following primary antibodies: YBX2 (MSY2) (1:2000, kindly provided by Dr. R.M. Schultz, Department of Biology, University of Pennsylvania, Philadelphia, PA), GATA1 (1:200, Santa Cruz, Heidelberg, Germany), GCNA (kindly provided by Dr. G. Enders, University of Kansas, Kansas City, KS), NOS1 (nNOS) (1:200, BD Biosciences, Heidelberg, Germany), TCFAP2C (Clone 6E4, 1:250, kindly provided by Dr. H. Hurst, Cancer Research UK, London, UK), for 1 h at 37°C except YBX2 (4°C, overnight). The secondary antibodies were anti-rabbit (1:500; DAKO, Hamburg, Germany) and anti-rat (1:200; DAKO) for 30 min at room temperature. Signals were visualized using Vectastain ABC Kit (Vector Laboratories, Axxora Deutchland, Lörrach, Germany). Sections were photographed using Diskus software (Diskus, Hilden, Germany). Files were processed using Adobe Photoshop Elements 5 and Illustrator software (Adobe).

Isolation and Immunostaining of Primordial Germ Cells

Embryos at E 7.5 were collected in Dulbecco modified Eagle medium (DMEM) (Gibco, Karlsruhe, Germany) supplemented with 0.5% BSA. Embryonic fragments from the base of the allantois were isolated using a microscalpel and incubated in 0.05% trypsin and 0.5 mM EDTA for 7 min, followed by dissociation into single cells by trituration with a mouth pipette. Dissociated single cells were pipetted onto poly-l-lysine-coated glass slides and fixed in 2% PFA. Cells were permeabilized in 0.1% Triton-X-100/PBS, and primary antibody incubation was carried out overnight at 4°C. The following antibodies and dilutions were used: anti-Tcfap2c (H-77, 1:100; Santa Cruz Biotechnology, Inc.), anti-Blimp-1 (Prdm-1 3H2-E8, 1:20; Affinity Bioreagents, Golden, CO), Alexa-488 (1:500; Invitrogen, Karlsruhe, Germany), and Alexa-594 (1:500; Invitrogen).

Detection of Cell Death

LysoTracker (Invitrogen, Germany) staining was performed as previously described [19]. Acidic compartments within apoptotic cells and regular cells engulfing debris from apoptotic cells are labeled. For these analyses, E 8.0 mouse embryos were prepared, incubated in LysoTracker solution, i.e., 5 μl LysoTracker, 1 ml DMEM (without phenol-red), and 2 mg/ml BSA (Sigma Aldrich, Munich, Germany), at 37°C for 40 min, washed twice in DMEM/BSA, twice in PBS, and fixed with 4% PFA in PBS for 2 h at 4°C. Whole mounts were incubated with 4′,6-diamidino-2-phenylindole (DAPI) (Sigma Aldrich), and the staining was analyzed using a Zeiss-ApoTome (Carl Zeiss) with a fluorescence module and appropriate filter set.

Differentiation and Selection of PGC from Embryoid Body Cultures

Essentially, the procedure was performed as described previously [20]. For embryoid body (EB) formation, ES cells were washed twice in calcium- and magnesium-free Dulbecco phosphate solution (DPBS, Invitrogen, Germany) and dissociated with 0.05% trypsin/EDTA solution (PAN-Biotech, Aidenbach, Germany). Trypsin was neutralized with DMEM (Invitrogen) containing 15% fetal bovine serum (Hyclone, Bonn, Germany). To remove feeder cells, the ES cells were plated twice on a 10-cm culture dish (TPP, Trasadingen, Switzerland) with 0.2% gelatin. After 45 min, nonadherent ES cells were collected, and hanging drop culture (750 ES cells per 25 μL) on lids of dishes filled with PBS was initiated. After 3 days in hanging drop culture, the resulting EBs were transferred to bacterial dishes (10–15 EB per dish) to prevent attachment. Every second day, one-third of the medium was replaced. SSEA-1 positive cell fractions were obtained using MACS-Beads and MACS-LS columns (Miltenyi Biotec, Cologne, Germany) according to the manufacturer's protocol. Five microliters anti-SSEA-1 antibody (0.1 mg/ml; R&D Systems, Heidelberg, Germany) per 2 × 10⁵ total cells were used.

Tcfap2c Is Expressed in Primordial Germ Cells and Gonocytes

First the expression of Tcfap2c in PGCs and gonocytes during embryogenesis was analyzed. Using immunohistochemistry, we detected Tcfap2c-positive cells starting from the early bud stage at E 7.25 (the earliest stage analyzed) at the base of the allantois, the region where PGCs are located at this developmental stage (Fig. 1, A and B, and Supplemental Fig. S1A). To confirm that the cells expressing Tcfap2c were germ cells, the base of the allantois (box in Fig. 1A) was dissected, trypsinized, and labeled with antibodies detecting the PGC markers Prdm1 and Tcfap2c (Fig. 1, C1 and C2). Prdm1 and Tcfap2c were coexpressed in these cells (Fig.1, C3 and C4), suggesting that Tcfap2c is expressed in PGCs. Next, we extracted RNA from the dissected allantoises at E 7.25. RT-PCR analyses showed that the markers for PGCs Kit and Dnd1 (previous symbol Ter) were coexpressed with Tcfap2c (Supplemental Fig. S1B). These results are in agreement with data obtained from transgenic mice expressing PRDM1-mVenus, where transgene expression was detected in Tcfap2c-positive cells at the base of the allantois in E 7.25 embryos [9]. Taken together, these data demonstrate that Tcfap2c is expressed in PGCs shortly after specification. At E 8.25, PGCs have begun to migrate into the embryo along the hindgut and express the marker Dppa3. At this stage, Tcfap2c expression (Fig. 1D) colocalizes with the signal obtained for Dppa3 (Fig. 1, E1–E3). Using in situ hybridization, we could detect Tcfap2c transcripts in the region of the hindgut at E 9.5. The observed pepperlike pattern is typical for migrating PGCs (Fig. 1F). At E 10.5 and E 11.5, we detected Tcfap2c transcripts in postmigratory PGCs in the genital ridges (Fig. 1, G and H). Staining of paraffin sections with a TCFAP2C-specific antibody confirmed this pattern from E 11.5 to E 12.5 (Fig. 1, I and K). TCFAP2C protein or RNA could neither be detected in germ cells in genital ridges after E 12.5 nor in adult testes or ovaries (Fig. 1L and Supplemental Fig. S1, C–E). These results are in agreement with data obtained from humans where expression of Tcfap2c was observed in primordial germ cells and gonocytes, but not in prespermatogonia and spermatogonia [15, 16]. These results prompted us to investigate the role of Tcfap2c in germ cell development.

FIG. 1.

Tcfap2c is expressed in primordial germ cells and gonocytes. Immunoflourescent detection of TCFAP2C at (A–C) E 7.25 and (D and E) E 8.25. Boxed area in A is enlarged in B. C1–C4) Double labeling of cell suspension generated by dissecting boxed area from A using (C1) PRDM1 and (C2) TCFAP2C antibodies; (C3) counterstaining with DAPI; and (C4) merge of C1–C3. D) Detection of TCFAP2C expression at E 8.25 along the forming hindgut. Boxed area in D is enlarged in E1 and E2 and shows the fluorescent signal for DPPA3 and TCFAP2C, respectively. E3) Merge of E1 and E2, demonstrating the coexpression of both markers. Messenger RNA in situ hybridization of Tcfap2c at (F) E 9.5, (G) E 10.5, and (H) E 11.5. Arrows in B, C1, C2, E3, F, and G indicate primordial germ cells. I–L) Antibody detecting TCFAP2C in sections at (I) E 11.5, (K) E 12.5, and (L) E 15.5. a, allantois; gr, genital ridge; lb, limb bud. Bar = 50 μm.

Tcfap2c Deficient Mice Fail to Reproduce

A complete loss of Tcfap2c is lethal to the embryo around E 6.5 due to a placenta defect and thus precludes the analysis of PGC development [12, 13]. To overcome this difficulty, we took advantage of transgenic mice bearing conditionally mutated Tfap2c alleles [21]. These Tcfap2c ^flox/flox mice [21] were bred to either Mox2-cre mice [22] expressing the Cre recombinase from E 6 on in the embryo proper or to Alpl-cre mice [23] expressing the Cre protein in PGCs. Both, Tcfap2c ^flox/flox:Mox2-cre and Tcfap2c ^flox/flox:Alpl-cre animals developed normally but failed to reproduce and displayed a similar phenotype with respect to germ cell development. Testes and ovaries of the mutant animals were dramatically reduced in size compared to wild-type controls (Fig. 2, A and B). In female animals, lack of germ cells in ovaries was demonstrated using YBX2 as marker for oocytes (Fig. 2, D vs. C). In male animals, atrophy of the seminiferous tubules was apparent (Fig. 2, F vs. E), and immunohistochemistry demonstrated that mutant testes did not contain any germ cells as judged by the marker YBX2 (Fig. 2, H vs. G) [24]. In contrast, GATA1-positive Sertoli cells were present (Fig. 2, K vs. I) [25] as well as Leydig cells that could be visualized with anti-NOS1 (Fig. 2, M vs. L) [26]. Of note, Leydig cell hyperplasia (Fig. 2, lch) was obvious in the mutant testis, which most likely is a secondary effect due to lack of germ cells. In some animals, the germ cell atrophy was incomplete. PCR analysis of laser microdissected tissue revealed that the Cre-mediated loss of Tcfap2c had not occurred in such testicular areas (Supplemental Fig. S2), suggesting that the incomplete atrophy was due to an incomplete Cre-mediated excision. Indeed, the recombination frequency of the Alpl-cre allele has been reported to reach 72% [27]. These results demonstrate that loss of Tcfap2c leads to sterility.

FIG. 2.

Loss of Tcfap2c leads to lack of germ cells. A) Testes and (B) ovaries of control (Wt) and Tcfap2c^flox/flox:Mox2-cre (mut) mice. Bar = 2 mm. MSY-2 immunohistochemistry on ovarian tissue of (C) control and (D) Tcfap2c^flox/flox:Mox2-cre. Sections of (C, E, G, I, L) control and (D, F, H, K, M) Tcfap2c^flox/flox:Alpl-cre mutants stained with hematoxylin and eosin (H&E) or the antibody indicated. Specimen and sections were prepared from 3-mo-old animals. st, seminiferous tubules; lch, Leydig cell hyperplasia. Bar = 200 μm (E–M) and 500 μm (C and D).

Tcfap2c Deficient Germ Cells Are Lost Shortly after Specification

Because the temporal expression of Tcfap2c in the germ cell lineage is restricted to PGCs and gonocytes, we tried to determine at which stage during early germ cell development might Tcfap2c be essential. To address this question, we crossed the Tcfap2c ^flox/flox mice with Sox2-cre transgenic mice. The Sox2-cre transgene is expressed from E 5.0 in all cells of the epiblast [28]. Therefore, Sox2-cre transgenic mice could be used in a manner similar to the Mox2-cre-transgenic mice to overcome the placenta defect caused by complete loss of Tcfap2c. However, Cre-mediated recombination is more efficient compared to Mox2-cre or Alpl-cre, thus, precluding a mosaic recombination within a tissue (unpublished data and [28]).

Germ cell development of Tfap2c ^flox/flox:Sox2-cre or control embryos was monitored using AP staining. PGCs could be detected at the primitive streak stage (E 7.25) at the base of the allantois both in the mutants and in the littermate controls (Fig. 3, A and B, arrows). The total number appeared not to be altered. However, at E 8.0, the numbers of PGCs in mutant mice were drastically reduced compared to control mice. In mutant animals, only a few PGCs were detectable near the base of the allantois, and none had initiated migration in contrast to the wild types (Fig. 3, D vs. C, arrows). At E 9.5, no migrating PGCs could be detected along the hindgut in mutant animals in contrast to controls (Fig. 3, E vs. F, arrows). Moreover, at E 10.5 the genital ridges of mutant animals lacked AP staining indicative of PGCs (Fig. 3, H vs. G, arrows,). At E 12.5, wild-type genital ridges contained large amounts of germ cells as indicated by the intense staining. In contrast, the genital ridges of Tcfap2c ^flox/flox: Sox2-cre mice were completely devoid of the AP signal (Fig. 3, J vs. I). Immunohistochemical staining using the germ cell marker GCNA at E 11.5 confirmed that mutant animals lacked germ cells (Fig. 3, K and L). These results demonstrate that the loss of Tcfap2cin PGCs interfered with early germ cell development from E 8.0 onward.

FIG. 3.

Tcfap2c is required for germ cells. Alkaline phosphatase (AP) whole mount staining of control (wt) and mutant (Tcfap2c^flox/flox:Sox2-cre, mut) (A and B) whole embryos, (C and H) trunk, and (I and J) genital ridge at the embryonic age indicated. The arrows indicate the position of AP-positive germ cells. K, L) Antibody to GCNA detection of germ cells in sections of genital ridges of E 11.5 control (WT) and mutant (mut). gr, genital ridge; kd, kidney; mn, mesonephros. Bar =100 μm (C–J) and 25 μm (K and L).

No Enhanced Apoptosis in Tcfap2c-Deficient Germ Cells

To investigate a possible loss of PGCs due to apoptosis, as observed in Pou5f1-deficient mice [29], LysoTracker analysis [19] of E 8.0 Tcfap2c ^flox/flox:Sox2-cre embryos was performed. LysoTracker signal in the region between the allantois and the developing hindgut, where PGCs are localized at this point of embryonic development, was not enhanced in the mutants (Fig. 4A) compared to wild types (Fig. 4B). Apoptosis in E 8.0 head-fold mesenchyme (Fig. 4C) was easily detected and served as positive control. The bright-field images were merged with the signal obtained by the LysoTracker to show the embryo and allantois (Fig. 4, A′–C′). Statistical analysis of this experiment showed that the differences between the numbers of LysoTracker positive cells in wild type (11 ± 2.55) compared to Tcfap2c-deficient (11.74 ± 2.73) embryos were not significantlt different ( Fig. 4D; P = 0.722, Student t-test). In addition, no difference in apoptosis could be detected using whole mount TUNEL-assays (data not shown) [30]. Therefore, cell loss by apoptosis is unlikely to be the mechanism responsible for the lack of germ cells in Tcfap2c mutant animals.

FIG. 4.

Cell death is not enhanced in mutant PGCs. Photomicrography of LysoTracker analysis (red signal, LT) of (A and A′) mutant (mut) and (B and B′) wild type (Wt) animals at E 8.0. The allantois (a) and posterior end of embryo is shown. C, C′) Apoptosis in the head mesenchyme of wild-type animal serves as positive control. A–C) Counterstaining with DAPI. A′–C′) Merge of bright field (BF) and lysotracker (red, LT) signal. Bar = 100 μm. D) Graph showing the number and standard deviation of cells positive in the posterior end of embryos (n = 8) in the LysoTracker study for wild type (WT; 11 ± 2.55) and Tcfap2c mutants (11.74 ± 2.73).

PGCs Derived from Tcfap2c-Deficient Embryoid Body Cultures Upregulate Somatic Markers

The defects observed in PGCs of Tcfap2c mutants are a phenocopy of the defects observed in Prdm1-deficient mice. Loss of Prdm1 leads to induction of somatic differentiation as demonstrated by the upregulation of Hoxb1 [8]. We used an in vitro model for PGC differentiation to test whether lack of Tcfap2c would likewise result in Hoxb1 gene activation. PGCs are known to be present in EB cultures derived from murine embryonic stem cells. Stage-specific embryonic antigen (SSEA1)-positive cells obtained from EB cultures represent candidate PGCs [20]. Tcfap2c-deficient ES cells were generated from Tcfap2c-deficient blastocysts derived from a Tcfap2c +/− line that was bred into the 129Sv background (our unpublished results). EB cultures were initiated from wild-type ES cells as well as from Tcfap2c-deficient ES cells. We used two different lines of Tcfap2c-deficient ES cells, yielding similar results. After 3, 5, and 7 days, the EBs were trypsinized, and SSEA1-positive cells (the nascent PGCs) were isolated using immunomagnetic bead sorting. As published [20], expression of Pou5f1 remained constant in SSEA1-positive cells while Pou5f1 levels declined in the SSEA1-negative population (data not shown). The markers for early PGCs, Prdm1 and Dppa3, could be detected in both wild-type and Tcfap2c-deficient cultures (Fig. 5A, top), indicating that PGCs had been specified in both genotypes. However, the markers for mature PGCs (Nanos3, Dazl, and Mutvh), which are also expressed in ES cells, decreased significantly in SSEA1-positive cells of Tcfap2c-deficient EB cultures compared to wild-type controls over the time observed (Fig. 5A, middle). Notably, the markers for somatic differentiation could either be detected only in cells from Tcfap2c-deficient EB cultures (Hoxb1) or were induced stronger and earlier in this assay (Hoxa1, T, and Snai1) (Fig. 5A, lower). Taken together, these data suggest, that in Tcfap2c-deficient EB cultures, PGCs are specified but further germ cell-specific differentiation is impaired and somatic differentiation is initiated.

FIG. 5.

Lack of Tcfap2c results in upregulation of somatic markers. A) Wild-type (Wt) and Tcfap2c-deficient (Tcfap2c −/−) ES cells were subjected to hanging drop cultures to induce embryoid body formation (EB) for 3 (d3), 5 (d5), and 7 (d7) days. SSEA1-positive cells were isolated after trypsinization, and RT-PCR analyses of expression of the genes indicated was performed. Primer sequences are given in Supplemental Table S1. The experiments were repeated 10 times. B) TCam-2 cells were treated with siRNA mixture to repress TCFAP2C (kd) and analyzed with respect to untreated cells (u) or cells treated with unspecific siRNA (scr). RT-PCR was used to detect the expression of the genes indicated; (−) indicates negative control without RT. Primer sequences are given in Supplemental Table S1. The experiments were repeated three times.

Loss of TCFAP2C Leads to Upregulation of Mesoderm Markers

Kurimoto et al. [9] demonstrated that Prdm1 regulates germ cell specification by the inhibition of somatic differentiation, specifically by repressing HOX cluster activation, epithelial-mesenchymal transition, cell cycle progression, and the DNA-methyltransferase machinery. Therefore, we analyzed the role of Tcfap2c in the genetic program orchestrated by Prdm1. As an in vitro model, we used TCam-2 [31], a cell line derived from a human seminoma that displays significant characteristics of PGC. Indeed, TCam-2 cells express several PGC markers such as PRDM1 and PRMT5 [32], and TCFAP2C, SOX17, KIT, DPPA3, NANOG, and POU5F1 [33–35]. In addition, TCam-2 cells exhibit symmetric dimethylation of arginine-3 on histones H2A and/or H4 tails (H2A/H4R3me2s) [32]. This modification has been demonstrated to be specific for PGCs and is directed by the interaction of Prdm1 and the arginine-specific histone methyltransferase Prmt5 [36].

First, we tested whether TCam-2 cells could serve as an in vitro model for PRDM1-mediated repression of the somatic program on the genetic and epigenetic level. Here, short interfering RNA (siRNA)-mediated knockdown of PRDM1 led to a reduction of TCFAP2C expression levels and a concomitant upregulation of HOXB1. In addition, expression of the PRDM1 target gene MYC was induced while the expression of NANOG remained unaffected (Supplemental Fig. S3A). Western blot analyses demonstrated that the symmetric dimethylation on histone H2A/H4 was lost after PRDM1 knockdown (Supplemental Fig. S3B). These data indicated that TCam-2 seminoma cells can serve as a model to study PRDM1-mediated gene programs in PGCs and further corroborated that TCFAP2C represents a PRDM1 target gene.

Next, we established a siRNA-mediated knockdown of TCFAP2C in TCam-2 cells to analyze TCFAP2C-dependent genetic programs in more detail. We compared the knockdown efficiency of three single siRNAs to their mixture (as suggested by the supplier) and detected the strongest inhibition of TCFAP2C RNA and protein when using the mixture (Supplemental Figure S3C). Next, we performed quantitative RT analyses of 31 markers (see Supplemental Table S1). As depicted in Supplemental Fig. 3D, expression of pluripotency genes (NANOG, POU5F1, SOX17, SOX2, UTF1, and REXO1), genes involved in DNA de novo methylation (DNMT3A, DNMT3B, and DNMT3L), signaling (SMAD4, SMAD7, and STAT5A), and various markers for ectoderm and endoderm differentiation remained unaffected after downregulation of TCFAP2C. However, the genes indicative of mesoderm differentiation, MYOD1 and HAND1, showed strong upregulation. Also GATA2, a marker found in mesoderm-derived hematopoietic progenitors [37], was upregulated. Using conventional RT-PCR, induction of MYOD1 and HAND1 as well as HOXA1 and HOXB1 could be detected in TCam-2. Induction of T was only weak compared to the results of the EB cultures while the levels of Snai1 were not altered in this context. Of note, levels of PRDM1 remained unaffected by TCFAP2C knockdown (Fig. 5B). Taken together, we demonstrated that repression of TCFAP2C leads to upregulation of HOXB1 and the mesoderm markers MYOD1 and HAND1 while markers of other germ layers, cell cycle progression, signaling, and DNA methyltransferases remain unaffected. These results therefore suggest that TCFAP2C is involved in the repression of a set of genes inducing mesodermal differentiation.

In summary we propose a model, wherein Tcfap2c is an effector of PRDM1-mediated repression of somatic differentiation, in particular differentiation into mesoderm. Pluripotency-associated markers and genes involved in epigenetic reprogramming are not affected.